-
Notifications
You must be signed in to change notification settings - Fork 195
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update CI matrix to use NVKS nodes. #3572
Merged
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
General allocation strategy is: - Primary CUB testing continues to use v100 (32GiB). This is because CUB tests often require very large amounts of gmem. - Other CUB builds use t4 (16GiB). These should have enough memory to run most tests. - Thrust testing uses t4 (16GiB). Some tests may require >8GiB, but not as much as CUB requires. - libcudacxx/cudax/python testing uses rtx2080 (8GiB), as these are not as memory intensive as Thrust/CUB. None of the NVKS queue require the testing tag anymore, so this has been removed as well.
7089df0
to
50864b9
Compare
This makes sure that they run in the correct CI config on appropriate hardware.
heterogeneous/barrier_abi_v2.pass.cpp is timing out on rtx2080.
🟨 CI finished in 5h 06m: Pass: 98%/166 | Total: 1d 03h | Avg: 10m 05s | Max: 2h 28m | Hits: 527%/22672
|
Project | |
---|---|
+/- | CCCL Infrastructure |
libcu++ | |
+/- | CUB |
Thrust | |
CUDA Experimental | |
python | |
CCCL C Parallel Library | |
Catch2Helper |
Modifications in project or dependencies?
Project | |
---|---|
+/- | CCCL Infrastructure |
+/- | libcu++ |
+/- | CUB |
+/- | Thrust |
+/- | CUDA Experimental |
+/- | python |
+/- | CCCL C Parallel Library |
+/- | Catch2Helper |
🏃 Runner counts (total jobs: 166)
# | Runner |
---|---|
118 | linux-amd64-cpu16 |
16 | windows-amd64-cpu16 |
10 | linux-amd64-gpu-rtx2080-latest-1 |
10 | linux-arm64-cpu16 |
10 | linux-amd64-gpu-t4-latest-1 |
1 | linux-amd64-gpu-h100-latest-1 |
1 | linux-amd64-gpu-v100-latest-1 |
miscco
approved these changes
Jan 29, 2025
bdice
reviewed
Jan 29, 2025
🟨 CI finished in 56m 44s: Pass: 98%/166 | Total: 1d 01h | Avg: 9m 22s | Max: 36m 08s | Hits: 528%/22672
|
Project | |
---|---|
+/- | CCCL Infrastructure |
libcu++ | |
+/- | CUB |
Thrust | |
CUDA Experimental | |
python | |
CCCL C Parallel Library | |
Catch2Helper |
Modifications in project or dependencies?
Project | |
---|---|
+/- | CCCL Infrastructure |
+/- | libcu++ |
+/- | CUB |
+/- | Thrust |
+/- | CUDA Experimental |
+/- | python |
+/- | CCCL C Parallel Library |
+/- | Catch2Helper |
🏃 Runner counts (total jobs: 166)
# | Runner |
---|---|
118 | linux-amd64-cpu16 |
16 | windows-amd64-cpu16 |
10 | linux-amd64-gpu-rtx2080-latest-1 |
10 | linux-arm64-cpu16 |
8 | linux-amd64-gpu-t4-latest-1 |
3 | linux-amd64-gpu-v100-latest-1 |
1 | linux-amd64-gpu-h100-latest-1 |
bernhardmgruber
approved these changes
Jan 29, 2025
This will allow us to reuse more build artifacts, and works around some issues with libcudacxx (NVIDIA#3590).
CUB -> RTXA6000 (48GiB) Thrust -> RTX4090 (24GiB) Others -> RTX2080 (8GiB)
🟩 CI finished in 38m 36s: Pass: 100%/156 | Total: 21h 46m | Avg: 8m 22s | Max: 30m 13s | Hits: 546%/21523
|
Project | |
---|---|
+/- | CCCL Infrastructure |
libcu++ | |
+/- | CUB |
Thrust | |
CUDA Experimental | |
python | |
CCCL C Parallel Library | |
Catch2Helper |
Modifications in project or dependencies?
Project | |
---|---|
+/- | CCCL Infrastructure |
+/- | libcu++ |
+/- | CUB |
+/- | Thrust |
+/- | CUDA Experimental |
+/- | python |
+/- | CCCL C Parallel Library |
+/- | Catch2Helper |
🏃 Runner counts (total jobs: 156)
# | Runner |
---|---|
110 | linux-amd64-cpu16 |
14 | windows-amd64-cpu16 |
12 | linux-amd64-gpu-rtx2080-latest-1 |
10 | linux-arm64-cpu16 |
6 | linux-amd64-gpu-rtxa6000-latest-1 |
3 | linux-amd64-gpu-rtx4090-latest-1 |
1 | linux-amd64-gpu-h100-latest-1 |
🟩 CI finished in 52m 52s: Pass: 100%/156 | Total: 22h 57m | Avg: 8m 49s | Max: 52m 32s | Hits: 546%/21523
|
Project | |
---|---|
+/- | CCCL Infrastructure |
libcu++ | |
+/- | CUB |
Thrust | |
CUDA Experimental | |
python | |
CCCL C Parallel Library | |
Catch2Helper |
Modifications in project or dependencies?
Project | |
---|---|
+/- | CCCL Infrastructure |
+/- | libcu++ |
+/- | CUB |
+/- | Thrust |
+/- | CUDA Experimental |
+/- | python |
+/- | CCCL C Parallel Library |
+/- | Catch2Helper |
🏃 Runner counts (total jobs: 156)
# | Runner |
---|---|
110 | linux-amd64-cpu16 |
14 | windows-amd64-cpu16 |
12 | linux-amd64-gpu-rtx2080-latest-1 |
10 | linux-arm64-cpu16 |
6 | linux-amd64-gpu-rtxa6000-latest-1 |
3 | linux-amd64-gpu-rtx4090-latest-1 |
1 | linux-amd64-gpu-h100-latest-1 |
🟩 CI finished in 39m 44s: Pass: 100%/156 | Total: 22h 40m | Avg: 8m 43s | Max: 38m 43s | Hits: 546%/21523
|
Project | |
---|---|
+/- | CCCL Infrastructure |
libcu++ | |
+/- | CUB |
Thrust | |
CUDA Experimental | |
python | |
CCCL C Parallel Library | |
Catch2Helper |
Modifications in project or dependencies?
Project | |
---|---|
+/- | CCCL Infrastructure |
+/- | libcu++ |
+/- | CUB |
+/- | Thrust |
+/- | CUDA Experimental |
+/- | python |
+/- | CCCL C Parallel Library |
+/- | Catch2Helper |
🏃 Runner counts (total jobs: 156)
# | Runner |
---|---|
110 | linux-amd64-cpu16 |
14 | windows-amd64-cpu16 |
12 | linux-amd64-gpu-rtx2080-latest-1 |
10 | linux-arm64-cpu16 |
6 | linux-amd64-gpu-rtxa6000-latest-1 |
3 | linux-amd64-gpu-rtx4090-latest-1 |
1 | linux-amd64-gpu-h100-latest-1 |
miscco
pushed a commit
that referenced
this pull request
Jan 30, 2025
* Update CI matrix to use NVKS nodes. * Update windows CI scripts to accept -arch. * Move all non-Catch2 device algo tests to lid0/lid1. This makes sure that they run in the correct CI config on appropriate hardware. * Switch to all rtx queues: CUB -> RTXA6000 (48GiB) Thrust -> RTX4090 (24GiB) Others -> RTX2080 (8GiB)
bernhardmgruber
pushed a commit
that referenced
this pull request
Jan 30, 2025
* Update CI matrix to use NVKS nodes. * Update windows CI scripts to accept -arch. * Move all non-Catch2 device algo tests to lid0/lid1. This makes sure that they run in the correct CI config on appropriate hardware. * Switch to all rtx queues: CUB -> RTXA6000 (48GiB) Thrust -> RTX4090 (24GiB) Others -> RTX2080 (8GiB) Co-authored-by: Allison Piper <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
General allocation strategy is:
None of the NVKS queue require the testing tag anymore, so this has been removed as well.
🏃 Runner counts (total jobs: 156)
linux-amd64-cpu16
windows-amd64-cpu16
linux-amd64-gpu-rtx2080-latest-1
linux-arm64-cpu16
linux-amd64-gpu-rtxa6000-latest-1
linux-amd64-gpu-rtx4090-latest-1
linux-amd64-gpu-h100-latest-1